JPS58151975A - Detecting method for fault of flash butt welding - Google Patents

Abstract

PURPOSE:To detect whether welding is normal or not, by arranging plural magnetic detecting elements dispersedly along the butt part between the rear end of a preceding steep plate and the front end of a trailing steel plate. CONSTITUTION:The magnetic detecting elements for magnetic field distribution detection are arranged dispersedly along the weld zone. Namely, when they are equipped to an electrode, grooves 60 are formed in the electrode 30 on the top surface side in a comb-tooth shape; tips (on a steel plate end side) of the teeth are made deep to arrange detection coil 7 therein, intermediate parts are made shallow to run lead lines 7a of the detection coils 7 therein, and the other-end common part is made slightly deeper to allow the lead lines to run therein. A magnetic field produced by a current flowing through the steel plate during flashing and upsetting is detected to find =>1 of a flashing frequency distribution, flashing mean current distribution, flashing heat input distribution, upsetting mean current distribution, and upsetting heat input distribution from respective output voltages of the detecting elements, thereby detecting whether the welding is normal or not.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for detecting the quality of welding results in flash butt welding of wide steel plates such as hot or cold rolled thin steel plates.

As is well known, the steel material that has been rolled into a thin sheet is wound into a coil, and the coil is sent to the next step for rolling it even thinner, or for pickling, plating, etc. However, since the length of the steel plate of the coil is finite, there will be interruptions in the processing, which is not preferable.

Therefore, in pickling, plating, etc., the beginning and end of each coil are welded to make them continuous, and recently, the same method has begun to be used in rolling, but in the former, since a large force is not applied, some welding is required. Although there is no problem even if there is a defect, in the latter case, large tension and compressive forces are applied, so if there is a defect in welding, it may lead to breakage, which may lead to a major accident.

Flash butt welding is used for this type of welding, and in this welding method, as shown in FIG.
A voltage is applied by the power supply 5 to generate a flash between the wire start and end, and when the wire start and end are sufficiently heated, they are pressed together and a large current (upset current) is passed for a short time to complete the welding. . 6 is the welded part.

An example of a flash butt welding device developed for the thin plate rolling process is shown in FIGS. 4 to 6. FIG. 4 is a front view, FIG. 5 is a plan view, and FIG. 6 is an enlarged view of a portion of FIG. 4, all of which show an outline. This welding equipment cuts the ends of the steel plates, flash-butt welds, and cuts off the welds.
is fixed at the exit side device, and 20 is the input side device, which is equipped with a trolley 21 and is movable. A welding power source (transformer) 5 is mounted on the exit device 20, and as shown in FIG.
5d, each having a pair of conducting wires 22.
23 (Fig. 4). A female member 24 of a tulip contact is attached to one end of each conductive wire 22, and these are respectively fitted into male members 25 (there is also a plurality of these) provided on the outlet device 10. One end of each conductive wire 23 is connected to a plurality of points on a rod-shaped upper electrode 4a extending in the width direction of the steel plate, and similarly, a plurality of points on the upper electrode 3a are connected via conductive wires 26 (there are also a plurality of them). and is connected to the male member 25. Similarly, the rod-shaped lower electrodes 4b and 3b extending in the width direction of the steel plates are not connected and simply serve to support the steel plates 1 and 2 (not shown in FIGS. 4 to 6).

A rotary shear is used to cut the end of the steel plate at right angles to the threading line so that the rear ends and tips of the leading and trailing steel plates can be brought into close contact with each other, and 30 is the blade of the shear of the entry side device. As shown in FIG. 6, this blade consists of four blades, upper sides 30a, 30b and lower sides 30c, 30d, and cuts the steel plate at two locations at the same time. Reference numeral 31 denotes a support portion of the rotary shear, which is movable along the rail 32 in the width direction of the steel plate. Although a rotary shear is also provided in the exit device 10,
Only the support portion 33 and rail 34 are shown in the figure.

A flash trimmer for cutting off the weld part is also attached to the exit device 10, and 35 is the upper trimmer and 36 is the lower trimmer, and the support part 33 also serves as a support part for the lower trimmer 36. 37 is a support portion of the upper trimmer 35, and 38 is its rail. 39 is a roller for conveying steel plates, 40.41
Although not shown, a steel plate holder is provided on the exit side device, and a steel plate holder is also provided on the input side device. 42 is a cross adjustment cylinder for centering, 43 is the same side guide, 4
4.45 is an image sensor for width setting, 46 to 50 are cylinders, 51.52 is a rod, 53 is a guide, 54 is a side guide and puncher, and 55 is a suction duct for dust and gas generated during welding.

When cutting the end of the preceding steel plate, the steel plate presser 46 is lowered by the cylinder 46 to clamp the steel plate together with the lower presser 41,
Stop moving. During this time, rolling in the next step continues, but since a looper (not shown) is provided next to this apparatus, the steel plate is fed through the louver. When the steel plate stops, the rotary shear of the outlet device is operated and the steel plate is cut at a position where the steel plate slightly protrudes toward the input side from the electrodes 3a and 3b. Next, when the leading edge of the trailing steel plate passes through the entry side device, the steel plate holder of the entry side device is activated to stop the steel plate, and then the rotary shear 30 is activated to cut the steel plate leading edge. After that, the cylinder 47 operates and the entrance device 2
0 in the direction of arrow F, and the tip of the trailing steel plate butts against the rear end of the leading steel plate. At this time, the female contact 24 fits into the male contact 25. In addition, as shown in FIG. 5, the male contact 25. (There are four, 25a to 25d.) can be moved forward and backward by a cylinder 48, thereby ensuring a secure fit with the female contact 24.

Next, the electrodes 3a, 3b+ 4a+ 4b are brought into contact with the steel plate, and when the power is turned on, the conductor 22, the contact 24.
25, a current flows through the path of the conductive wire 26, the electrode 3a, the #i4 plate, the electrode 4a, and the conductive wire 23, and a flash occurs at the steel plate butt portion. The entry device 20 is slowly fed by the cylinder 47, and the flash continues, and after a predetermined time, the absent cylinders 49, 50 (Fig. 5) are activated, and the rod 51, 52 moves the entry device 20 to the exit side. It is pulled into the device 10 and the abutting portions of the leading and trailing steel plates are crimped. At this time, a large current (upset current) flows, further heating the abutted portion, and when the current is cut off, welding is completed.

Generally, the welded portion bulges out a little on both the upper and lower surfaces, but this is removed by flash trimmers 35 and 36.

In such steel plate flash butt welding, the width of the steel plate is 1
．． It is wide, around 5 m, and it is not easy to perform flash butt welding uniformly over the entire width. In addition, all operations have been automated, and equipment has become more complex, larger, and faster, making it difficult for operators to access the welding area. It is virtually impossible to judge the welding state sensually.

On the other hand, if a coil breaks during continuous rolling due to poor welding, the rolling rolls will be damaged, and the line will have to be shut down to change rolls and polish the rolls, which will reduce production volume, yield, and This has a large impact on costs. Therefore, we can automatically detect whether the Franchebutt weld is good or bad.
If it is defective, it is desirable to be able to take appropriate measures such as immediately cutting off the welded portion and re-doing the welding.

Japanese Patent Application Laid-Open No. 54-90 for determining the quality of flash welding
038r flash weld quality determination method and device”
, JP-A-54-90039r, "Method and apparatus for determining quality of flash welded parts" have been proposed. These calculate the power consumption from the voltage between welding electrodes and the total welding current,
This method attempts to judge the quality of welding based on whether or not a predetermined amount of power has been applied, but in this method, if the welded part can be considered a spot, it may be difficult to weld if the welded part has a very large angle, length, and/or area. It cannot be guaranteed that all welds are uniformly welded. In other words, in the case of steel plates, defects may occur due to the flatness and/or straightness of the plate end surface when cutting the edges, changes in plate thickness in the width direction, uneven plate material, etc., and poor contact between the electrode and the plate due to foreign matter getting caught. , Due to non-uniformity, the welding current tends to be non-uniform in the sheet width direction, which leads to non-uniform flushing and upset heat input, which in turn leads to non-uniform fusion, distortion, etc. of the welded part, and the welding strength is partially insufficient. Although it is easy to make a weld, it is difficult to judge whether or not a defective weld has occurred just by checking the total heat input.

The present invention has been made in view of the above problems, and is intended to detect the magnetic field at a plurality of points along the entire length of a welded part, and to detect the weld length and defects from the detected output. In other words, when the current flowing through the plate becomes non-uniform, the magnetic field generated around the plate also becomes non-uniform as shown in Figure 2, and this can be detected by arranging a plurality of magnetic detection elements 7 as shown in Figure 3. By doing so, it is possible to know the current distribution in the plate, and in turn, it is possible to determine the weld length and defects. In addition, in Figure 2, a, b,
c indicates the part where the current is concentrated, and the magnetic flux φ generated by the current flowing through the plate 1.2 becomes a loop that surrounds the shelf wire plate, but a magnetic flux that surrounds only the part where the current density is high is also generated. However, the magnetic field strength distribution around the plate becomes non-uniform. The above determination can be made by detecting this with a magnetic detection element such as a coil, magnetoresistive element, or Hall pepper.

That is, the present invention provides a method for detecting the quality of flash butt welding between the terminal end of a leading steel plate and the tip of a trailing steel plate, in which a plurality of magnetic detection elements are distributed and arranged along the abutting portion between the terminal end and the leading edge, and And a detection output of the magnetic field created by the current flowing through the steel plate during upset is generated, and from each output voltage of these detection elements, the flushing frequency distribution, flushing average current distribution, flushing heat input distribution, upset average current distribution, and upset person are determined. The present invention is characterized by determining one or more of the heat distributions and detecting the quality of the welding result based on the results.This will be further explained below with reference to a specific example.

Magnetic detection elements for detecting magnetic field distribution may be distributed along the welded portion, but FIG. 7 shows an example in which a detection coil is attached to the electrode 3 (or 4). A comb-like tooth 1160 is formed on the surface side of the electrode 3, and as shown in FIG. is placed, the lead wire 7a of the detection coil 7 passes through the shallow middle part, and the common part (bone part) of the other end is shallow.
The lead wires of each detection coil pass through it deeply. Note that this figure does not show the terminal portion for supplying welding current.

FIG. 7 (C1 shows the external appearance of the detection coil 7. The detection coil is placed in the electrode groove so that its axis, that is, the center line of the bobbin winding body, faces the longitudinal direction of the electrode 3. As is clear, the magnetic field created by the current flowing through the steel plate can be detected.

FIG. 3 shows an example in which this electrode is the upper electrode 3a closest to the outlet device 10. 61 is an electrode holder, 62 is a cable connected to the lead wire of the detection coil, 63 is an amplifier, 64
is a data recorder, and 65 is an electromagnetic controller. Electrode 3
Since a large number of detection coils, for example, 11, are provided in a, the amplifier 63, data recorder 64, etc. are required for 11 channels. Note that recording may be performed by appropriately thinning out the data, and in the experiment described later, 7 channels were used.The data recorder 64 also records the truck speed S1, the total current i, and the welding voltage e. 66, 67, and 68 are isolation amplifiers.

Figures 9 and 10 show examples of measurement results. In these figures, 71 to 77 indicate the detection coil numbers, 76 is the coil in the center of the electrode, 71 to 75 are the coils on the work side, and 77 is the coil number on the work side. This is the coil on the drive side. As shown in the figure, when the upset is entered, the total current i becomes a large current,
The detection output voltage of the coil also increases. In the flushing process, these pulses have small amplitudes but change minutely, and sharp pulses appear here and there as shown by arrows. These pulses indicate flashing (sparking) occurrences. In this example, the width of the transverse plate is smaller than the electrode length, so that the detection coils 71, 72 extend outside the steel plate. Therefore, the outputs of the detection coils 71 and 72 do not increase even if the signal enters absent.

Figure 10 shows the measurement results when an insulating paper is placed on the surface of the steel plate directly below the center detection coil to prevent contact with the electrode.This figure shows that even if the sensor enters the absent, the output of the detection coil 76 is was not very large, and it can be assumed that the flash butt welding at the center of the steel plate was defective (cold welding) (which was the case). Note that if the area of the insulating paper is too small, there will be little effect. In the example shown in FIG. 10, 30×12 cm insulating paper was used, and the width of the horizontal board was 921 mm. The following can be determined from the detection coil output during flushing and absent.

(1) Flushing process: The output of each detection coil at the time of flushing is led to the comparison circuit, and the output of each detection coil is extracted by comparing it with the one above the threshold value by "'-5" to extract the flashing, and it is input to the counter and counted, and the total of each counter is calculated. Since the value indicates the number of times of flushing of the butt part of the steel plate part where the detection coil is located, the distribution of the number of times of flushing of the butt part can be determined from the counter count value, and whether or not the flushing occurs uniformly or not, and whether the flushing process was normal or not. etc. By full-wave rectifying and smoothing the flushing signal, that is, the component above the threshold of each detection coil, the average flushing current distribution can be found.
Furthermore, since the flushing heat input distribution can be found by full-wave rectification and integration of the flushing signal, it is also useful to use these to determine whether the flushing process is normal or abnormal.

(2) Upsetting process: By full-wave rectification, smoothing, and integration of the upset current, the average upset current distribution and heat input distribution can be determined, and from this it can be determined whether or not the upsetting process was normal.

(3) Poor board end shape: If the number of times of flushing and the average current distribution determined in (1) above are uneven at the beginning but normal thereafter, it can be determined that the shape of the cut end surface of the board is slightly defective (some edge shape (Defects may melt and disappear during flushing.) Mild shape defects do not adversely affect welding results that much. If the heat input distribution determined in (1) above is considerably non-uniform, the shape of the plate end face is quite poor, often resulting in poor welding. Said (2)
If the upset average current and heat input distribution determined by are non-uniform, the shape of the plate end face is extremely poor, but this is also related to the subsequent entrapment of foreign matter or inclusion.

(4) Foreign object jamming: If a foreign object (such as the above-mentioned insulating paper) is caught between the electrode and the steel plate, the absent distribution described in (2) above becomes non-uniform, and it is possible to estimate foreign object jamming from this.

Flushing uses a small current, so even if there is a foreign object, it will be maintained by the circulating current from the surrounding area and will not be affected much. However, if the flushing is uneven, the foreign object will be quite large and the current will be concentrated locally. There is a risk of electrode damage, so it is best to issue a mountain pass warning immediately.

(5) Steel plate inclusions: Inclusions (foreign objects) may be mixed into the steel plate, and in this case, the flashing becomes uneven and the upset distribution suddenly becomes uneven, and from this it is assumed that there are steel plate inclusions. can.

In this way, multiple detection coils are placed along the weld zone to detect the magnetic field distribution created by the current flowing through the steel plate during the flushing and upsetting periods, and to detect the flushing frequency distribution,
By measuring the flushing average current distribution, flushing heat input distribution, upset average current distribution, and absent input distribution, it is possible to detect various abnormalities in flushing butt welding of wide parts, which is extremely effective for preventing weld breakage and maintaining quality. be.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of flash butt welding of steel plates, Fig. 2 is an explanatory diagram of the magnetic field distribution around the steel plate, Fig. 3 is an explanatory diagram of the arrangement of detection coils, and Figs. A schematic front view and a plan view of an example of a flash butt welding device, FIG. 6 is a partially enlarged view of FIG. 4, and FIG.
(bl and (C1 are plan views, cross-sectional views, and coil perspective views of the electrodes, FIG. 8 is a block diagram showing the configuration of the detection signal processing circuit, and FIG. 9 and FIG. 1θ are waveform diagrams showing actual measurement results. In the drawing, 1 is a leading steel plate, 2 is a trailing steel plate, and 7 is a magnetic detection element. Applicant: Minoru Aoyagi, Patent Attorney, Nippon Steel Corporation

Claims (1)

[Claims] In a method for detecting the quality of flash butt welding between the terminal end of a leading steel plate and the tip of a trailing steel plate, a plurality of magnetic detection elements are distributed along the abutting portion between the terminal end and the leading edge, and the detection method The detection output of the magnetic field created by the current flowing through the steel plate is generated, and from each output voltage of these detection elements, the flashing frequency distribution, flashing average current distribution, flushing heat input distribution, absent average current distribution, and absent human heat distribution are calculated. seek one or more,
A method for detecting the quality of flash butt welding, which is characterized by detecting the quality of the welding result.